expectreg (version 0.16)

expectreg-package: Expectile Regression

Description

Expectile and quantile regression of models with nonlinear effects e.g. spatial, random, ridge using least asymmetric weighed squares / absolutes as well as boosting; also supplies expectiles for common distributions.

Details

ll{ Package: expectreg Type: Package Version: 0.16 Date: 2010-09-07 License: GPL (>= 2) LazyLoad: yes }

• This package requires the packagesBayesX,mboostandsplines.

References

Koenker R (2005) Quantile Regression Cambridge University Press, New York Schnabel S and Eilers P (2009) Optimal expectile smoothing Computational Statistics and Data Analysis, 53:4168-4177 Fenske N and Kneib T and Hothorn T (2009) Identifying Risk Factors for Severe Childhood Malnutrition by Boosting Additive Quantile Regression Technical Report 052, University of Munich He X (1997) Quantile Curves without Crossing The American Statistician, 51(2):186-192

See Also

mboost, BayesX

Examples

data(dutchboys)
## Expectile Regression using the restricted approach
expreg <- expectile.restricted(dutchboys[,3] ~ base(dutchboys[,2],"pspline"),smooth="schall")
## The calculation of expectiles for given distributions
enorm(0.1)